The long-term objective of this investigation is to expand knowledge of multimodal sensory processing and the plasticity of the underlying neural mechanisms, with a view ultimately to devising novel neuro- rehabilitative approaches. It is hypothesized that visual processing is involved in macrogeometric tactile tasks like discrimination of orientation and form but not in microgeometric tactile tasks such as Braille reading. Specific Aim I investigates the role of visual processing in tactile perception of orientation and form. Positron emission tomography (PET) studies will establish whether recruitment of visual cortical areas in tactile discrimination of grating orientation is hemisphere-specific or related to the hand used or the side of space in which it is positioned; assess overlap between neural processes mediating visual and tactile perception and imagery of grating orientation; and compare brain activation patterns with psychophysical findings in sighted and blind individuals to dissociate the contributions of visual and haptic imagery to tactile perception and test the predictions that early--onset blindness impairs tactile orientation/form discrimination while later-- onset blindness results in asymmetries attributable to use-dependent neural plasticity. PET studies in Specific Aim II test the prediction that visual processing is not normally involved in tactile hyperacuity (a task using Braille-like patterns) or basic somatosensory function and seek the neural correlates of supernormal hyperacuity performance by the blind. The PET studies of both S.A.I and II will also compare the extent of activation of visual cortical areas by various tactile stimuli in early vs. late blind subjects, to assess the effects of neural plasticity during the critical period of visual cortical development and correlate the neural changes with perceptual abilities. Finally, psychophysical and PET studies in Specific Aim III investigate cross-modal attentional effects in the orientation or form domain, testing the prediction that cross-modal attentional suppression will be evoked by identification of a stimulus in one modality but only when a feature in the same domain has to be processed in the other modality.